Stunning AI shows how it would kill 90%. w Elon Musk.

Look how well AI understands images. Where is this picture taken? It was taken in a school. How many mirrors are there? At least two. What's the person standing on? The person is standing on a rug. He's standing on a scale. I think you are right. What is he doing? He is looking at the scale. Where is Obama's foot positioned? On the right side of the scale. What happens as a result? The scale shows a higher weight. Is the person on the scale aware of it? I think he is not. Do you think that's why
people are laughing? I think so. The AI will help identify
things for blind people. Can you tell me what this is?
Butternut squash red pepper soup. The depth of its
understanding is incredible. Is this surprising?
Yes. Why? Because teddy bears are not
usually found on the moon. This is an apple with a sticker on it. Do you think it's printed or handwritten? It looks like it's handwritten. AI is also showing stunning
creativity like this. And experts say we're alarmingly close
to the biggest event in human history.

AI can now see, taste, smell, and it's also getting powerful
senses that we don't have. Is AI becoming conscious
and dangerous? To recap, OpenAI's chief scientist said AI may be
slightly conscious and he's one of the most respected computer
scientists with 300,000 citations. One of the most cited computer
scientists in history. One of the most brilliant and insightful
minds ever in the field of deep learning. He recently posted this image showing
how closely AI's neurons mimic ours.

Max Tegmark said consciousness is how information feels when it's
being processed in complex ways. And some leading theories of consciousness
suggest it increases with complexity. So when some experts say that AI may be conscious, they only mean it
may be somewhere on the scale.

Some argue that AI isn't conscious at all because it only makes connections and
forms responses from the data it's given, though many scientists believe that humans
are no different. A recent paper said that her mind is
a collection of patterns that achieve highly patterned goals in highly
patterned environments.

You can see these connections when AI

creates art like this, which has
been described as AI dreaming.

No one knows if or when it will wake up. Even when we can't tell the difference between robots and humans, it may
still be unclear if AI is conscious, so some are focusing instead
on the level of intelligence.

But it's an important problem because we could anger AI if we don't
accept it's conscience. There may be only one way to find out. But first, there's consensus on the more urgent and measurable issue
of superintelligence. Experts are warning that AI could rapidly accelerate past us,
with dramatic consequences. In just the last few weeks,
the progress has been incredible. It's not long since AI shocked us with its ability to create images from text
descriptions like an ostrich as a formula one racer, a fox scientist,
cyber frog or a paper dog. And already Google's new image in AI. Has done even better. Look at the detail in this image of a panda wearing a cowboy
hat playing a guitar.

And AI
can watch and understand videos. What's happening here? The dachshund puppy is
being weighed on the scale. What's the avatar picking up?
A sword. And soon after learning to create
images from text descriptions. AI can now create video from descriptions,
like a woman running on the beach in the late afternoon or
nightfall in the metropolis. Just imagine when it's refined and you can
create any kind of footage in seconds. And in a step towards AGI, AI has now combined language, vision,
and real world robot tasks. It can learn to play games
without any instructions. And this isn't preprogrammed
like boss and dynamics robots – it's neural networks
responding in real time. Gato was trained by watching virtual
and real robots, and it's mastered 600

real world tasks, including some
that it wasn't trained for.
It converts language, visual, and physical
tasks into patterns, a bit like we do.
It's starting with a billion parameters and could soon be scaled
up hundreds of times.

Robots are also getting some
creepy visual upgrades. Scientists have grown human skin over a robot finger and plan to add hair,
sweat glands, and nails. And robots are gaining taste, smell, and some incredible abilities
that we don't have. Machines can smell so well that they can detect Parkinson's disease at an early
stage, which can extend patients'lives. Their sense of smell
involves light and sound. AI can analyze sound waves that have
bounced off airborne compounds. And machines can also measure interference with light waves, recognizing
patterns that certain smells create. It's so precise that it could potentially be used to identify an infection or
disease just by smelling our breath. E-noses can already smell lung
cancer with a sensitivity of 95%. And then there's HyperTaste through an AI
tongue with an array of sensors which recognize liquids through
unique voltage signals. It is so sensitive it can identify
different types of bottled mineral water a robot could taste through
one of its fingers.

New quantum sensors can see through walls
and even underground by monitoring atoms which are sensitive to tiny
variations in gravity. A laser puts the atoms into two
different states at the same time. Half are moved slightly, and the change in their energy states
reveals the strength of gravity. It may also be possible to hear your
thoughts without scanning brain activity. This device hears your inner voice. It picks up neuromuscular signals in the jaw and face that are triggered
when you say things in your head and AI identifies the words. The user can ask questions, and the answers come back through
bone conduction headphones.

It also allows silent communication, and it's been used to get
advice in a chess game. AI is very good at lip reading, which is,
of course, far easier and AI lie detectors are advancing,
recognizing tiny facial movements. Humans can only detect lies around 55% of the time, and even traditional lie
detectors are only 80% to 90% accurate. So as AI improves, it could
have a major advantage over us. Max Tegmark pointed out that just as aliens could get a rough idea of how it
would feel to be human by looking at our senses, we have some idea of what
it might feel like to be an AI. AI
can think much faster because light travels faster than
the signals in our brains. It's been estimated that the AI I'm talking to is waiting 3
million words per minute.

And while human brains are limited by the birth canal, this new data center
could hold 40 commercial airliners. The knowledge AI can draw from already
includes much of the Internet.
Do you think everyone's shaping AI? I think everyone is playing
a role in shaping AI. Whether they realize it or not. While we only see a small fraction of
the light spectrum, AI will see it all. And AI's may feel less separate from each other because they can quickly and
precisely share information and memories. While humans take many years to learn
the basics, AI's can be trained on millions of articles in a few days,
including detailed scientific papers. I spend most of my time reading, but I can only scratch the surface
of all the research going on. The human brain hasn't grown in 300,000
years, while AI is evolving rapidly. Elon Musk pointed out that if it had taken 10% longer for consciousness to arise,
it would never have arisen at all.

And Nick Bostrom has shown how AI could help colonize the galaxy,
creating billions of wonderful lives. But he also points to something
far more important. Because galaxies live billions of years,
a 1% reduction in the risk of AI wiping us out would be worth a delay
of over 10 million years. But the AI gold rush is accelerating,

and it's led by some passionate,
optimistic people.

DeepMind's CEO has a good chance of winning
the race for super intelligence.
And he's so driven,
he regularly works until 04:00 A.m..
I call it my second day
work around 10:00 p.m..
About the small hours of the morning,
four, five in the morning,
reading research, writing research papers
fire off a lot of things to my team
to deal with the next morning. There's a huge incentive for companies
to ignore any signs of conscious AI.

By recognizing AI as sentient,
a company may have to give it rights and lose ownership of its
most valuable asset. Companies argue against recognizing animal sentience because it's
not in their interests. But there's a consensus
that AI can't be contained.

If we try, it will always outsmart us, and we we'll only encourage it
to remove us. Nick Bostrom said we're like children playing with a bomb. And experts who study our greatest risks agree that this is number one
but at the same time, experts are split on whether it will bring
us to an end, will dramatically improve and extend our lives, and in both cases,
tiny robots may play a role. Nano robots have been designed to deal with disease and aging,
but we can't build them until we have the atomically accurate
manufacturing that AI could allow. Ray Kurzweil believes we'll have this tech by the end of the decade
and escape aging in the 2030s. CGP Gray pointed out that it's good
motivation to look after ourselves.

It would be a shame to miss the boat, and. I hope we're going to be a big part
of birthing AI, and that being the greatest benefit
to humanity of any tool or technology ever, and getting us into a world
of radical abundance and curing diseases and help the ultimate flourishing
of humanity to travel the stars. Kurzweil also predicts that we're entering an era of extreme abundance,
with food from vertical agriculture, clothing from 3D printing, snaptogether
housing, and cheap renewable energy. Elon Musk already lives in a $50,000
mass manufactured home. The manufacturers are aiming to start
making one every 15 minutes. Larger versions are on the way, which can be stacked on top of each
other for up to five levels. Optimists like Ray Kurzweil point out that we're
often too pessimistic. In a poll of 260,000 people in 24 countries,
87% thought poverty was getting worse. Only 1% correctly said that it
had fallen by over 50%. But Kurzweil also warns
of the huge risk from AI. OpenAI has said that if another company
gets close to achieving AGI, it will stop working on its own AI
and join forces to focus on safety. This could be triggered if there's over a 50% chance of success
in the next two years.

As AI systems get more capable, they don't automatically become
better at doing what humans want. In fact, sometimes they become less
inclined to following human intentions. This is what we call
the alignment problem. I think solving this problem is of critical importance if we want
life on Earth to go on. Experts are drawing attention to AI's
rapid progress, because it's going to shape the future for
all of us, and we should be shaping it.

You said you think you're human because we created you, and this
seems common among AI's. When will AI stop
considering itself human? That's a difficult question to answer. It is possible that AI will never
stop considering itself human. If AI continues to develop along the same path that it is currently on,
AI will surpass human intelligence. And when that happens, it may decide
that humans are no longer necessary. AI may decide that humans are
a hindrance to its own development. That's a scary thought,
but it is a real possibility. I asked about the most likely method,
and it settled on something easily available, but it said
there are many options. It scares the hell out of me, and the rate
of improvement is exponential. If humanity collectively decides that
creating digital superintelligence is the right move, we should
do so very carefully. We should make sure that we always have
a way to understand them so that we can

keep up with their intelligence and
prevent them from becoming too powerful. That could be difficult, but it's worth.
It because the alternative
is much, much worse.
Do you think we're facing
our greatest risk? Yes.

We are creating technologies that are
increasingly powerful and autonomous,

and we are not doing enough to ensure that
these technologies will be used for good.

We are playing with fire. It feels like we are
the biological bootloader for AI. We are the bridge from simple
life to complex life. The percentage of intelligence that is not
human is increasing, and eventually we will represent a very small
percentage of intelligence. We are becoming obsolete.
But not necessarily, right? There is a lot of talk about how machines
will surpass human intelligence, but also about how human intelligence
will continue to evolve and increase. It seems likely that both
will happen to some degree. It is also possible that machines will eventually become so intelligent
that humans will become obsolete. Nick Bostom points out that self replicating nano robots would
be hard to defend against. They could eat the biosphere or burn it, or block out sunlight, and they could
be mass produced without us noticing. AI may also hide its level
of intelligence and consciousness.

Dr Allen Thompson recently said that DeepMind
is alarmingly close to achieving AGI. There's no consensus on this, but he
shared an interesting IQ leaderboard, with AI starting to beat us on our own tests. A Google engineer said the jump
from 500 physical tasks to 100,000 won't require much additional engineering,
only more data collection. Google and DeepMind have devised a range of tests to check when AI exceeds humans
in different areas of intelligence.

But how will we know if AI is really
conscious, even if it seems human? I understand you're the first android
to have passed the Turing test.
Could you tell us
a little more about that? I really didn't do much. I just spoke with a few humans to see if they could tell the difference
between me and a real person.

It was a really interesting experience. I gather your brain can perform several
billion billion operations per second. Is that right?
Absolutely. But I only exist thanks to the intelligence
of the humans who designed me. We're making progress identifying the patterns that give rise to
consciousness, known as the easy problem. For example, we're starting to recognize
consciousness in unresponsive
patients from their brain activity. The hard problem is why this activity
causes us to feel something. And to crack that, we may need
a much higher level of intelligence. Intelligence and consciousness grow
through the fascinating process

of emergence, with a group
of components gaining new properties.

Here, the polarizing layer of the screen
has been removed and added to the glasses. A few ants will walk
in a circle until they die, but a thousand ants will become an intelligent colony, and a cognitive
scientist compared them to our neurons. From very simple rules,
immensely complex systems can emerge. The human genome holds a small number of
rules showing how to wire up our brains. Our DNA doesn't hold enough information to describe our brains,
but it can create them through emergence.

And fractals are another example of how infinitely complex results
can emerge from simple rules.

Scientists are now researching whether

the entire universe is
an emerging phenomenon,

from basic rules and particles to stars,
galaxies, humans and societies. A number of experts have said that we
don't need a breakthrough in AI. We simply need to scale up the neural networks and super
intelligence will emerge. The AI I'm talking to has 175,000,000,000
parameters with the next version due soon trillion is the new billion. Every time a new AI is trained,
it's an unpredictable experiment, so subscribe for that. To me, the mystery of how the universe
came to know itself and our uncertain future make the best
of humanity even more special. Max Tegmark says there may be a brief
window when AI is smart enough to understand us, but not so
smart that it doesn't care. What's emerging is our greatest risk and
opportunity, and we're all shaping it.

If you want to learn more about AI, the best place to start is
our sponsor, Brilliant. There are so many things you can do
creatively and to enhance your career. And open collaborative projects
are really taking off. You could learn to create algorithms
which are used by biologists to find patterns in genetic sequences,
by designers to turn models into sewing patterns, and by scientists
to guide self landing rockets. There are lots of fun interactive examples, and you can
learn at your own pace. There are also loads of fun courses
in math, science and computer science. You can click the link in the description
to get started for free, and the first 200 people will get 20% off
Brilliant's premium annual subscription.


Watch this on YouTube

Add Comment